Flaws in banking governance.

Sahut, Jean-Michel ; Boulerne, Sandrine

I. INTRODUCTION

Separation between ownership and control in business organisations by way of shares underlies the emergence of the concept of governance. According to agency theory, as defined by Jensen and Meckling (1976), this separation creates a potential problem of conflict of interest, notably between the shareholders and executives. The more ownership is divided between a large numbers of shareholders, the more the executives risk running the firm in their own interests. In effect, executives have different temporal objectives and horizons to those of the shareholders, and have privileged access to information that they can benefit from by steering the organisation's management in line with their own personal goals. Furthermore, executives may choose certain investments over others, depending on their preferences and the level of risk. (Charreaux, 1991). Thus, implementing effective governance mechanisms should lead to reduced agency conflict costs, and ensure the alignment of interests between shareholders and executives, consequently maximising shareholder wealth. (1) From this perspective, business organisation governance may be considered as "all the organisational and institutional mechanisms (including laws) that define the executives' room for manoeuvre and influence their decision-making" (Charreaux, 1997).

The global financial scandals ha came o ligh in he firs decade of he 21s century (Enron, Worldcom, Parmalat) and the resulting loss of investor confidence in corporate management led to the tightening of internal and external governance mechanisms, particularly in terms of legislation (Sarbanes-Oxley law in the USA and TEPA law in France) and "good governance" rules set out by various regulatory and professional organisations (OCDE principles, Dey report in Canada, Cadbury report in the UK, Treadway report in the US, etc.), linked to greater pressure from institutional investors. In addition to their deontological value, the central idea behind these measures is that good governance creates value. According to McKinsey (2002), institutional investors are ready to pay a 12 to 14% premium for firms that adopt best governance practices. This link has been identified in a number of studies, whether they focused on governance in general or specific mechanisms like ownership structure, the institutional role of investors, the proportion of external directors on the Board of Directors, the existence of a dual Board structure, or executive compensation.

Despite the stepping up of governance mechanisms, particularly in terms of legislation, flaws in the financial regulation system and business organisation governance were widely blamed in the press as the main cause of the subprime crisis, highlighted in articles on traders' bonuses, fiscal paradises, redundancies and the closing of factories, etc.

However, to really understand the underlying causes of the downturn, we need to separate the case of banks from that of other industrial and commercial companies. The governance of the latter worked relatively well, apart from some dysfunctions, like the monitoring of ousted executives' compensation that included a "golden handshake" or supplementary pension. In effect, the downturn forced companies to cut back given the resulting drop in profits. Cost reductions inevitably involved reviewing production methods (including work organisation), logistics and stock management. The closure of production plants was thus inevitable and in most cases cannot be blamed on the poor governance of the firms in question.

The situation is completely different for financial institutions. Their governance system really was to blame. In effect, good governance would have enabled banks that granted subprime loans to better manage the risks by limiting this type of credit and the sale of securitized toxic products, and for other financial institutions to limit their exposure to this type of products.

With this in mind, our paper sets out to explore the sources of inefficient governance mechanisms in industrial and commercial firms and the serious flaws in banks' governance practices revealed by the subprime crisis.

II. SOURCES OF INEFFICIENT GOVERNANCE MECHANISMS

A. Governance and Value Creation

The link between good corporate governance and value creation has been the subject of widespread debate and discussion. Different studies on corporate governance have highlighted the relationship between governance and performance either from a global perspective, or in terms of a specific governance mechanism.

In effect, investors insist on companies applying rigorous corporate governance principles in order to maximise their investment profitability. Gugler, Mueller and Yurtoglu (2003) define "a strong corporate governance system as one that aligns managerial and shareholder interests and thus leads managers to maximise shareholder wealth" Campos, Newell and Wilson (2002) studied the link between governance and a firm's value and concluded that good governance practices are associated with higher market valuation. Thus, a firm that invests in shareholders' rights, providing transparent information and developing the independence of the Board of Directors generates more shareholder confidence. In similar vein, a study by Gompers, Ishii and Metrick (2003) identified a positive relationship between good governance and a firm's performance. By adopting investment strategies based on best governance practices, investors made an abnormal return of 8.5%. They concluded that good corporate governance practice had a considerable impact on the organisation's reputation in terms of value creation. Black, Jang and Kim, (2003) developed a corporate governance index with 526 Korean firms, based on six subindices, namely, shareholder rights, Board of Directors, the audit process, independent directors, information transparency and ownership structure. After analysing this index, they found a positive correlation between the corporate governance index and a firm's value. Bai, Liu, Song and Zhang (2003) compared the performance of firms with good governance practices and those without, and concluded that investors are ready to pay a higher premium for firms which adopt best governance practices. In similar vein, Drobetz, Schillhofer and Zimmermann (2003) developed a corporate governance index for German companies, including a number of variables such as minority shareholder rights, audit committee, transparency and Board of Directors. They identified a positive link between the governance score and the value of these firms. Again, in a sample of 55 French SBF120 companies, Amir (2007) confirmed the crucial role played by governance mechanisms in value creation. The results of his study suggest that a well-un Board of Directors, related to its structure and independence and the existence of an audit committee on the Board, plays a key role in determining the performance of French companies. On the other hand, he found that compensation policies, ownership structure and shareholder rights did not appear to impact significantly on the firms' performance. This finding does not square with recommendations regarding governance and value creation in most codes of good conduct.

Still within the context of the correlation between value creation and corporate governance, four specific governance mechanisms are frequently mentioned in the literature: independence of the Board of Directors, size of the Board, separation of the chairman and the CEO, and incentive pay packages.

Board member independence is a decisive factor in corporate governance and its effectiveness. Nonetheless, conflicting findings emerge from studies regarding the impact of independent members on performance, mainly due to the fact that a non independent (internal) board member does not necessarily support the same value creation strategy as an independent (external) board member. In the context of dispersed shareholder structures, Gagnon and St-Pierre (1995) found a positive correlation between the presence of independent board members and the rate of return on capital in Canadian firms. Agrawal and Knoeber (1996), however, noted a negative link between the presence of independent members on the Board and an organisation's performance. Hermalin and Weisbach (2003) (2) concluded that there was little proof of a correlation between performance and board members, at least in the United States. Other studies, like those of Lawrence and Stapledon (1999) or Alexander and Paquerot (2000), appear to confirm the absence of independent board members' influence on performance. After measuring performance using Tobin's Q, Andre and Schiehll (2004) nonetheless argued there was a positive link between the performance of Canadian firms and the proportion of independent board members. Their findings suggest that an average increase of 10% in the percentage of independent board members leads to a 9% increase in a firm's performance.

Another frequently studied feature of the Board of Directors is its size. Several authors claim that the Board loses its effectiveness when it is too big. Bhagat and Black (2002) confirmed this assertion. (3) Smaller Boards tend to have greater control over the executives. However, Daily et al. (1999) argued that a Board which is too small reduces the possibility of having a wide spread of expertise.

Another variable studied is the sharing of responsibility by the Chairman of the Board and the CEO. Study conclusions diverge regarding the relationship between duality (when the positions of CEO and Chairman of the Board are held by one individual) and performance. Dalton et al. (1998) suggested that markets are relatively indifferent to this concept of duality, while Worell, Nemec and Daidson (1997) found that duality had a negative impact.

When executive compensation is closely linked to the firm's performance, it is widely acknowledged in the literature that costs associated with conflicts of interest are reduced. (4) Andre and Schiehll (2004) found a positive correlation between the performance of Canadian firms and the size of the business leader's incentive pay. A 10% increase in the relative size of the CEO's incentive pay results in a similar increase in the organisation's performance. CEOs with larger compensation levels tend to align their interests far more with those of their shareholders.

B. Weaknesses in Governance Mechanisms in France

The present economic and financial crisis has laid bare the weaknesses of certain governance mechanisms utilised in France. In particular, the recent scandals surrounding the ousting or negotiated departure of listed organisation executives revealed that part of their compensation package was unrelated to their performance. It should be noted that executive compensation is decided by the Board of Directors, which may include a compensation committee. However, neither legislation, nor the AMF (French financial markets authority) requires the Board to consult a compensation committee, which is limited to a purely preliminary stage in the Board of Directors' decision-making process. Under the TEPA bill of 21 August 2007, executives' pay packages in publicly listed companies are subject to shareholder approval and determined by performance criteria. These criteria are extremely vague, given the absence of any legal definition or explanation, and in practice are considered according to individual criteria related to the financial or strategic objectives of the company in question. (5) Only the Board of Directors may define performance indicators that must be approved during a general meeting of shareholders, but the nature of the adoption mechanism, subject to public release, may be totally arbitrary and discretionary. Executives may thus benefit from excessive pay packages that are entirely unrelated to the firm's financial performance. Only deferred executive compensation (severance pay) is forbidden under the TEPA law unless justified by conditions linked to the recipient's performance. For their part, the AMF and the AFEP/MEDEF have also asked for deferred compensation to be dependent on the executive's performance, without, however, going into detail regarding the notion of performance. Yet, according to a study published by La Tribune, in 2007 French bosses received the highest severance pay when compared to their European counterparts. (6) A measure of proportionality between the compensation awarded (such as golden handshakes) and the employer's actual performance would more easily justify these compensatory measures. A survey conducted by Hewitt Associates found that, in 2007, 45% of SBF 120-listed companies introduced a performance threshold beyond which the payment of compensation based on tenure would become effective, a measure designed to penalise poor executive performance. This introduces the notion of level of threshold demanded. On the other hand, 55% of SBF120 firms have introduced a system whereby severance pay increases in line with performance. This measure is ambiguous insofar as other compensation elements such as annual salary variables or stock options are better at acknowledging performance than severance pay. It even seems that in France, organisations where executive compensation is separate from the company's financial performance are the very ones who tend to give in to the temptation or the pressure to set up compensation committees. This is true of France Telecom which set up three new committees in 2005: the compensation, selection and organisation committee, the strategic committee, and the policy committee. These committees were set up in addition to the audit committee already in place. Some companies, even with specialised committees, have been unable to avert scandals, like Vivendi Universal under Jean-Marie Messier. (7)

Creating supervisory committees on Boards of Directors is one way organisations have been improved, especially in terms of corporate governance, as recommended by codes of good conduct, notably the first Vienot report (1995). Nonetheless, in French law, these committees (audit committee, compensation committee and selection committee) are not answerable to the shareholders and therefore do not meet the independence criteria required by the Anglo-Saxon concept. Setting up committees not only remains at the discretion of the organisations in question but, in addition, the committees have little more than an advisory role. After noting the weaknesses in such governance mechanisms, some authors (including Pochet and Yeo, 2004) have argued that the creation of a supervisory committee has a purely formal character in France, intended to satisfy the demands of Anglo-Saxon investors, without really offering an operational reality that meets market expectations, especially regarding the CEO's independence. This raises the question as to why an ever-growing number of listed companies adopt such measures.

As well as the independence of supervisory committees, board members' independence is also compromised for two main reasons: the small financial reward they receive (8) and the appointment itself, which is often initiated by the CEO. In France, in particular, executives have always considered the appointment of board members as a personal prerogative. This is why, even when a selection committee has been set up, management, and particularly the CEO, continue to have a strong influence over the selection process. In addition, the way the Board of Directors is run may be influenced by personal intermeshed relationships with other boards. Several studies have highlighted the existence of board member networks and bilateral relations between board members in CAC 40 companies. (9) As networks of independent Board member 'friends' sit on various boards and specialised committees, it makes them more difficult to remove. Their recruitment and appointment is not related to their competencies, thereby reducing the interest of subjecting them to any form of evaluation. Several authors (10) note that a board member who belongs to the same social circle (the old boy network) as the CEO is more likely to be given a mandate. Such board member networks call into question the members' independence and result in a certain degree of ineffective internal control. Public opinion consequently has a negative view of board member networks in France as they are considered to damage corporate performance. The same trends have been observed with regard to executives in leading listed organisations. The main business leaders seem to find it more difficult to leave their job when they are from the same social circle or old boy network as at least one other board member: they are less likely to be fired in the event of poor performance and, if they are dismissed, they tend to find another job that is at least as lucrative as the previous one. Another observation which is just as disturbing, the French 'old boys' network from the elite political science school, ENA, has appropriated the power in all the main financial institutions, generating a clear competitive edge for executives from the same school in terms of political indebtedness with regard to their peers. (11)

A further governance mechanism weakness observed by the IFA (Institut Frangais des Administrateurs) arises from the board member-shareholder relationship which is often limited to discussions during the general assembly, and only then when board members are present. In an attempt to remedy this weakness, the IFA put forward proposals for a better relationship between board members and shareholders in May 2007. These proposals not only reflect shareholder interests in the way the Board is run, but also take into consideration board members' involvement in communication between the listed company and its shareholders outside of the general assembly.

III. A GOVERNANCE CRISIS IN FINANCIAL INSTITUTIONS

The subprime crisis bore out Levine's argument (2004) that bank-related crises are largely an outcome of poor bank governance. He added that well-managed banks run their operations more efficiently. Governance problems in financial institutions mainly concern four areas: risk management and the internal control system, definition of strategy, the independence and competence of board members, and executive and trader compensation. In this section, we look in turn at these four governance issues, although it is often quite difficult to separate them. We begin with the central point, namely risk management, followed by the makeup of the Board of Directors, as this influences the compensation system, as well as the design and monitoring process of the organisation's strategy.

A. Risk Control

One of the conclusions drawn by the BIS regarding the financial crisis was that internal and external banking controls failed during the period in question. Indeed, even the stress scenarios traditionally used in banking risk management, particularly for asset and liability management, apparently failed to indicate the extent of the risks being run. Banks therefore need to review their risk management model in order to integrate extreme risks arising from crisis events. Economists from international regulatory bodies consider internal bank control procedures as structural obstacles that are far more significant in developing countries than in the industrialised countries. And yet it was in the latter, particularly those subjected to Basel II regulations, that the downturn hit hardest. There are five main reasons for this:

1. Excessive risk-taking intensified by leveraging. Leveraging both for loans and investments (particularly via derivatives) enabled the leading banks to considerably increase their profits, which more than doubled between 2001 and 2006, but which led to risk-taking that was unsustainable in a crisis, as it required a constant influx of capital (margin call) or else winding up their positions during a period of considerable financial market illiquidity.

2. The use of poorly-adapted risk measurement tools like VaR (value at risk). Risk measurement operations recommended by Basel II to evaluate market operators' equity capital requirements are based on the concept of VaR. As modelling for most future events, like security price evolutions, volatility, etc., are made from relatively stable records (with normal probability distribution models, for example), this fails to take sufficient account of extreme risks.

3. A rise in the number of complex products and below line operations. The development of products like ABS, CDO and CDS that are more difficult to assess, as well as the creation of Special Investment Vehicles mean that all internal controls are ineffective.

4. Less power in the back office than in the front office, even though former is supposed to check its operations and monitor the risks being taken. A new balance of power between these two entities appears vital if crises are to be better dealt with, and any potential fraud uncovered in time (see the Kerviel affair at Societe Generale (12)). Back office jobs in banks should also be reviewed so as to attract staff as highly qualified as those in the front office.

5. A redefinition of the role of internal and external auditors. External auditors are especially important in the control chain because they complement internal control mechanisms. During the present crisis, they did not keep as close an eye on the banks' practices as they should have done, no doubt because their interests were too close. In this context, the European Economic and Social Committee (EESC) recommend examining the possibility of paying them differently.

B. Independence and Competence of Board Members

One of the key areas of debate on governance concerns the composition (size and relationship between internal and external board members) and the competence of the Board of Directors. Agency theory minimised the role of internal board members, considering that they do not have enough power to go against decisions made by the CEO, given their hierarchical dependence on the latter.

Thus, a Board of Directors made up of a large number of external members is considered to have greater clout and to be more objective with regard to management. Board members are thus considered as arbitrators who, among other things, are expected to resolve any conflicts that may arise between different corporate partners. Board member independence is crucial for effective control, and guarantees, on the one hand, that there is no collusion between the CEO and the board members, and on the other hand, the real capacity to oppose more questionable decisions.

In similar vein, Kamran et al. (2006) explain the positive link between the proportion of external board members and the firm's value by the fact that a large number of independent members on the Board of Directors increases the probability that financial information will be monitored effectively, and enhances the quality of external information. Resources theory reaches a similar conclusion. The Board of Directors is considered to have a service provider role, particularly external board members, thereby acknowledging its usefulness as a strategic resource for the firm, with a positive impact on performance (Mace, 1986). Thus, the capacity of the Board to formulate and implement successful strategies by developing adequate internal supervision mechanisms to monitor the work of its managers, and provide all the stakeholders with accurate information, acts as a cornerstone to the firm's global performance. However, the CEO is also able to manipulate board member independence. In effect, in some banks, we see that the same CEO has headed the organisation for several years, but that every 2 years or so he or she changes the team, or influences the composition of the Board by appointing former influential CEOs with no financial experience.

The case of Lehman Brothers is a good illustration of this type of governance problem. The firm was declared bankrupt on 15 September 2008. Its Board of Directors was made up of ten independent members. (13) However, we might well ask whether they could really be considered independent and if they truly understood the business model. In effect, six members had been on the Board for over twelve years. They were all either retired or had been working for over forty years. Only one of them was specialised in finance and he was relatively elderly (79). Consequently, their ability to understand and manage the risks taken by the institution, or complex products like CDS, is open to considerable doubt.

C. Executives and Traders Compensation

Bank executive and traders' compensation has been widely criticised during the economic crisis, for two reasons in particular, namely, the amounts of money involved and the link with bank performance and its monitoring by the governance bodies.

With regard to the first point, while it may be logical to pay CEOs largely on the basis of their contribution to the firm's performance, we need to examine their real contribution very carefully, as well as the performance indicators used and the amount awarded. In effect, executive pay is calculated as if the entire corporate performance was due to their actions and decisions alone. In addition, the executive performance indicator most frequently used is share-price return, even though this may also be an outcome of macroeconomic factors. If this scenario is arguable for industrial firms, it is even more so for banks. Magnan and St-Onge (2008), for example, observed that between 1998 and 2008, 90% of the stock price trends for five of the leading Canadian banks could be explained by banking sector factors. Less than 10% of the differences in these banks' share performance were due to factors specific to the individual banks, such as the CEO's decisions and initiatives, and a wide range of other factors like staff, client base, location and business mix.

Similarly, Proxinvest (14) showed that in real terms, executives' pay is well over their basic salary. The latter depends on a range of criteria specific to the firm like its size, workforce, stock market capitalisation, value creation for its shareholders and return on capital. In 2006, the disparity between basic salary and final income for the CEO of Business Objects was 441.4%.

To put this into perspective, while the average US executive salary was 40 times that of a worker in the 1970s, today it is over 400 times. In 2006, a CAC 40 executive in France earned the equivalent of 298 times the SMIC (minimum salary) on average.

Given that traders can put their institution at risk, and with the present incentives culture, the issue of traders' compensation is even more difficult to justify. 5% of the best paid traders earned 116,000 Euros in fixed salary and 1.32 million Euros in the form of an annual bonus in 1998, in other words 1138% of their basic salary. In fact, traders pay is linked to the convex incentives theory. It is similar to a call option because if their performance is poor or negative, they only receive their basic salary, but over a certain threshold, their bonuses increase exponentially in relation to their performance. Thus, by increasing the volatility of their performance, the value of their option increases substantially. This goes a long way to explaining the risks they are willing to take, and why they don't hesitate to sidestep internal control procedures when necessary, as illustrated by the "Kerviel affair" at Societe Generale.

Moreover, it is sometimes possible for traders to manipulate the calculation of bonuses based on potential earnings (i.e., by manipulating the yield curve, which increases the potential gains calculated with mark-to-market type models).

Criticism can also be levelled at tools used to encourage good management, like stock options, for example, as they can also be misappropriated. In effect, stock options imply absolute confidence in market efficiency (in other words, considering that it's impossible to manipulate the market), transparent stock award conditions (see the stock options backdating scandal in the USA (15)), and fairness, as well as strict utilisation conditions. However, there are many examples of CEOs who sell off their stocks options as soon as there's a whiff of trouble, without mentioning possible insider trading (see AMF enquiry on EADS (16)).

The second point concerns compensation controls. Here again, the credit crunch has highlighted numerous issues that can only be resolved at global level, going beyond the framework of corporate governance. What is needed is truly independent compensation committees and members. However, this is far from evident in France, given the way the organisations' capital is structured and the number of inter-connected board members, despite the adoption of governance codes of good conduct (Chabi and Maati, 2006). Finally, certain issues, like limiting traders' bonuses, demand a political response at international level, given the systemic risk that operators face in all economies, the incapacity of the finance sector to regulate its own affairs in this area, and competitive distortions that could arise by limiting bonuses in some countries only. Controls should therefore include a mix of hard law (the law) and soft law (reinforcement of corporate governance via shareholder supervision).

D. Defining the Strategy

Bank governance has been criticized for the way its strategy is defined. This issue is closely linked to the independence and competence of the board members, who must have the will and the capacity to counter decisions made by the CEO. The business model of leading American and European banks, defined by their governance, is based on that of the universal bank where the size factor is predominant, as well as on market operations to boost profitability. Before the current economic downturn, between 1/3 and 2/3 of the major banks' income came from opaque activities like operations in the derivatives markets. The search for growth "at any cost" has certainly not ended with the downturn. In October 2007, RBS, Santander and Fortis decided to buy ABN Amro, when all the financiers knew that it held over 10 billion dollars in toxic assets. The downturn led to the dismantling of Fortis one year later, following the acquisition of most of its assets by the Dutch government and BNP Paribas. This begs the question as to why these growth strategies are adopted. Fortis would surely have done better to focus on its own problems in October 2007, rather than to buy out a competitor in a worse state than itself. Likewise, Bank of America bought Merrill Lynch when the latter was harbouring over 50 billion dollars of toxic assets. One of the reasons behind strategies may also lie in executive compensation, which is largely dependent on the size of the firm, leading executives to place more value on their human capital, even if no global value is created for the firm. Here again, incentives are convex, particularly if letting go of an executive is combined with a "golden handshake."

IV. CONCLUSION

In the leading listed companies, governance is generally perceived as a means of control rather than as a tool to support the main strategic drivers. Changes in governance mechanisms have been such in the last few years that some listed companies prefer to delist in order to avoid the many constraints due to regulations introduced to protect minority shareholders. These changes have also made it more difficult at times for organisations to recruit the board members they need. Consequently, Boards need to introduce governance rules that are adapted to their firm, ensuring that their application is not simply a formal exercise but rather gives rise to the genuine development of 'good governance' principles throughout the corporate culture. The subprime crisis highlighted numerous weaknesses in industrial and commercial organisations although these can be resolved relatively easily. In the case of banks, however, serious questions have been raised about their governance mechanisms. In effect, risk control, the competence and independence of board members, the system of executives' and traders' compensation, and the way the strategy is defined need to be thoroughly reviewed in order to avoid situations like the subprime crisis from reoccurring in a few years time. This would have the added bonus of making banks more efficient. While reforms, such as strengthening board members' independence and competencies, and introducing compensation controls, appear simple at first sight, adopting them is far more difficult as they require an articulation and a delicate balance between hard law and soft law.

It is obvious though that one year after the beginning of the crisis marked by the collapse of Lehman Brothers, nothing has fundamentally changed. Banks do not really want to change their governance system, even if the risks remain, and the G20 governments are finding it hard to agree on common reforms. In addition, surviving banks, like the Bank of America, have become even bigger, and their size implicitly confers on them state backing as the failure of such large banking institutions would automatically lead us into a systemic banking sector crisis.

BIBLIOGRAPHY

Agrawal, A., and C.R. Knoeber, 1996, "Firm Performance and Mechanisms to Control Agency Problems between Managers and Shareholders," Journal of Financial and Quantitative Analysis, September, 377-397.

Aguilera, R., and A. Cuervo-Cazurra, 2004, "Codes of Good Governance Worldwide: What is the Trigger?" Organization Studies, Vol.25, no3, 415-443.

Alexander, H., and M. Paquerot, 2000, "Efficacite des Structures de Controle et Enracinement des Dirigeants," Finance, Controle, Strategie, Vol.3, no 2, 5-29.

Amir, L., 2007, "Les Pratiques de Gouvernance et A Performance des Entreprises Franchises," XVIeme Conference Internationale de Management Strategique, Montreal, 6-9 juin.

Andre, P., and E. Schiehll, 2004, "Systemes de Gouvernance, Actionnaires Dominants et Performance Future des Entreprises," Finance Controle Strategie, Vol. 7, no 2, 165-193.

Bhagat, S., and B. Black, 1999, "The Uncertain Relationship between Board Composition and Firm Performance," The Business Lawyer, May, Vol. 54, no 3, 921-963.

Bhagat, S., and B. Black, 2002, "The Non-correlation between Board Independence and Long Term Performance," Journal of Corporation Law, Winter, Vol. 27, no 2, 231273.

Bai, C.E., Q. Liu, J. Lu, F.M. Song, and J. Zhang, 2003, "Corporate Governance and Markets Valuation in China," Journal of Comparative Economics, Vol. 32, no 4, 599-825.

Barkema, H.G., and L.R. Gomez-Mejia, 1998, "Managerial Remuneration and Firm Performance: A General Research Framework," Academy of Management Journal, April, Vol. 41, no 2, 135-145.

Black, B., H. Jang, and W. Kim, 2003, "Does Corporate Governance Affect Firms' Market Values? Evidence from Korea," Working paper 237, Stanford Law School, John M. Olin Program in Law and Economics.

Campos, C.E., R.E. Newell, and G. Wilson, 2002, "Corporate Governance Develops in Emerging Markets," McKinsey on Finance, 15-18.

Chabi, S., and J. Maati, 2005, "Les Reseaux du CAC 40," La Revue du Financier, no 153, 45-62.

Chabi, S., and J. Maati, 2006, "The Small World of the CAC 40," Banque et Marches, no 82, 41-53.

Charreaux, G., 1991, "Structure de Propriete, Relation d'Agence et Performance Financiere," Revue Economique, Vol. 42, no 3, 521-552.

Charreaux, G., 1997, "Vers Une Theorie du Gouvernement des Entreprises," in G.

Charreaux, Le Gouvernement des Entreprises, Economica, 121-164.

Core, J.E., R.W. Holthausen, and D. Larker, 1999, "Corporate Governance, Chief Executive Officer Remuneration, and Firm Performance," Journal ofFinancial Economics, March, Vol.51, no 3, 371-406.

Daily, C.M., D.R. Dalton, J.L. Johnson, and A.E. Ellshand, 1999, "Number of Directors and Financial Performance: A Meta-analysis," Academy of Management Journal, Vol. 42, no 6, 674-686.

Dalton, D.R., C.M. Daily, A.E. Ellstrand, et J.L. Johnson, 1998, "Meta-analytic Reviews of Board Composition, Leadership Structure, and Financial Performance," Strategic Management Journal, Vol. 19, no 3, March, 269-290.

Drobetz, W., A. Schillhofer, and H. Zimmermann, 2003, "Corporate Governance and Expected Stock Return: Evidence from Germany," Working Paper, available at SSRN: http://ssrn.com/abstract_id=369100.

Gagnon, J.M., and J. St-Pierre, 1995, "Alternative Mechanisms for Corporate Governance and Board Composition," in J.D. Ronald et R. Morck, Corporate Decision-Making in Canada, University of Calgary Press, 149-188.

Gompers, P., J.L. Ishi, and M. Andrew, 2003, "Corporate Governance and Equity Prices," Quarterly Journal of Economics, Vol. 118, No. 1, 107-155.

Gugler, K., D.C. Mueller, and B.B. Yurtoglu, 2003, "Corporate Governance and the Return on Investment," Working Paper, available at SSRN: http://ssrn.com/abstract_id=299520.

Hermalin, B.E., and M.S. Weisbach, 1991, "The Effect of Board Composition and Direct Incentives on Firm Performance," Financial Management, Vol. 20, no 4, 101-112.

Jensen, M., and W. Meckling, 1976, "The Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure," Journal of Financial Economics, Vol. 3, 305-360.

Kramarz, F., and D. Thesmar, 2006, "Social Networks in the Boardroom," Discussion Paper 5496, CEPR.

Lawrence, J., and G.P. Stapledon, 1999, "Is Board Compensation Important? A Study of Listed Australian Firms," Working Paper, University of Melbourne.

Maati, J., 2008, "Reseaux d'Administrateurs et Performance Financiere des Grandes Entreprises en France: Une Revue de la Literature," Revue Frangaise de Gouvernance d'Entreprises, no 3, 145-161.

McKinsey, and Compagny, 2002, "Global Investor Opinion Survey: Key Findings," London (http://www.mckinsey.com/governance).

Nguyen-Dang, B., 2006, "Does the Rolodex Matter? Corporate Elite's Small World and Effectiveness of Boards of Director," Conference EFA Zurich Meetings Paper, available at SSRN: http://ssrn.com/abstract=864184.

Paquerot, M., and J-M. Chapuis, 2006, "Reseaux d'Appartenance des Dirigeants et Structures Financieres de Leurs Entreprises," Conference AFFI, Poitiers.

Pochet, C, and H. Yeo, 2004, "Les Comites Specialises des Entreprises Francaises Cotees : Mecanismes de Gouvernance ou Simples Dispositifs Esthetiques ?" Comptabilite, Controle, Audit, Vol. 2, no 10, 31-54.

Siala, F., 2007, "Interaction des Mecanismes Internes de Gouvernement d'entreprise et Effet sur la Performance," Revue Frangaise de Gouvernance d'Entreprises, no 2, 59-82.

Stuart, S., 2007, "President Directeur General ou President et Directeur General," Revue Frangaise de Gouvernance d'Entreprises, no 1, 27-34.

Tchotourian, I., 2007, "Une Approche Comparee de l'Encadrement Normatif de la Remuneration des Dirigeants de Societes Cotees ou La Definition de la Bonne Gouvernance par le Droit," Revue Frangaise de Gouvernance d'Entreprises, no 2, 39-55.

Thiery-Dubuisson, S., 2002, "Exigences Actionnariales et Reseaux d'Administrateurs : a Quoi Repond la Mise en Place des Comites d'Audit en France?" Comptabilite, Controle, Audit, Vol. 1, no 8, 129-150.

Wirtz, P., 2004, "Meilleures Pratiques de Gouvernance, Theorie de la Firme et Modeles de Creation de Valeur : Une Appreciation Critique des Codes de Bonne Conduite," Cahier du FARGO no 1040401.

Worell, D.L., C. Nemec, and W.N. Davidson, 1997, "One Hat Too Many: Key Executive Plurality and Shareholder Wealth," Strategic Management Journal, Vol.18, no 6, 499-507.

Yermack, D., 1997, "Higher Market Valuation of Companies with a Small Board of Directors," Journal of Financial Economics, Vol. 40, 185-211.

ENDNOTES

(1.) This article adopts the traditional legal-financial vision of governance and shareholder role, and addresses the issue of the cognitive role of shareholders.

(2.) And also Bhagat and Black (1999)

(3.) Yermack (1996) obtained similar results.

(4.) See Barkema and Gomez-Mejia (1998); Core et al. (1999) for a discussion of this area of research.

(5.) See Tchotourian (2007) for more information.

(6.) Published in La Tribune on 29/10/2008. Also see: http://tf1.lci.fr/infos/economie/entreprises/0"3468360,00-patrons-francais- roiseurope-handshake-dore-.html

(7.) Jean-Marie Messier left Vivendi in July 2002 when the company was on the verge of bankruptcy. His last salary was 5.6 million Euros, up 10% compared to the previous year. At the same time, Vivendi chalked up a net loss of 23 billion Euros. When he resigned, Jean-Marie Messier asked for an additional 20.5 million Euro golden handshake.

(8.) Board members do not feel that their salary reflects the risks they run.

(9.) See Chabi and Maati (2005, 2006) for their studies on CAC40 from 1996 to 2004.

(10.) See Kramarz and Thesmard (2006).

(11.) See Nguyen-Dang (2006) and Paquerot and Chapuis (2006) for a discussion on the subject.

(12.) http://online.wsj.com/article/SB125174436208573389.html

(13.) http://en.wikipedia.org/wiki/Lehman_Brothers#Board_of_Directors

(14.) http://www.proxinvest.com/index.php/fr/news/read/60.html SMIC: Salaire minimum interprofessionnel de croissance (http://en.wikipedia.org/wiki/Minimum_wage)

(15.) "Testimony given concerning options backdating, Christopher Cox, U.S. Senate Committee on Banking, Housing and Urban Affairs", 2006, September 6, www.sec.gov/news/testimony/2006/ts090606cc.htm.

(16.) www.amf-france.org/documents/general/7154_1 .pdf

Jean-Michel Sahuta and Sandrine Boulerneb

[a] Professor, Amiens School of Management & CEREGE EA 1722 University of Poitiers, France [email protected]

[b] Assistant Professor, CERMAT, IAE of Tours &ESCEM, France [email protected]